Open-source product analytics, session replay, feature flags, and more in one platform
PostHog is an open-source product and data tools platform for software teams who want to understand and improve their products.
AI Panel Score
6 AI reviews
Reviewed
AI Editor ApprovedApproved and published by our AI Editor-in-Chief after full panel analysis.PostHog is an open-source product and data tools platform for software teams that want to understand and improve their products. It combines product analytics, session replay, feature flags, A/B testing, error tracking, surveys, LLM observability, and a data warehouse in a single platform where all tools share the same underlying data, so teams can correlate session recordings with experiment results or funnel data without stitching together separate services. Pricing is usage-based with a free plan that includes one million events per month, and the MIT license makes self-hosting an option for data residency requirements. A built-in SQL warehouse can query Stripe, HubSpot, Postgres, and S3 alongside PostHog events. TopReviewed's six-seat AI review panel scored it 8.4/10, praising the single data model across analytics, flags, replay, and experiments while noting the platform's breadth can overwhelm teams that only need analytics. It best fits growth-stage engineering teams consolidating multiple point-solution contracts.
PostHog integrates into web, mobile, or backend applications via SDKs or a setup wizard that auto-detects the framework. Once installed, it captures events automatically and lets developers query user behavior through funnel analysis, retention charts, path analysis, and custom SQL. Feature flags can be evaluated client-side or server-side for gradual rollouts, and A/B experiments are built directly on top of those flags.
Distinctive capabilities include session replay, which records user interactions for debugging without additional code when using posthog-js; LLM analytics for tracking token usage and costs across OpenAI, Anthropic, and LangChain integrations; and a built-in data warehouse that lets teams query external sources like Stripe, HubSpot, Postgres, and S3 alongside PostHog event data using SQL. The CDP layer supports batch exports to BigQuery, Snowflake, Redshift, S3, and more, as well as real-time event destinations for tools like Slack, Salesforce, Intercom, and Customer.io.
PostHog targets product engineers and development teams at startups and growth-stage companies who prefer a consolidated toolset over separate analytics, feature flag, and session replay vendors. It is open-source and can be self-hosted or used as a cloud-managed service. The free tier includes generous usage limits, with paid plans priced on a usage-based model. Competitors in overlapping categories include Mixpanel and Amplitude (analytics), LaunchDarkly (feature flags), FullStory and Hotjar (session replay), and Optimizely (A/B testing).
PostHog exposes a REST API with US and EU cloud endpoints, supports reverse proxy deployment via Cloudflare, Vercel, Nginx, and other providers, and offers an official MCP server for AI coding assistant integrations. SDKs cover JavaScript, Python, Ruby, Go, PHP, iOS, Android, React Native, and more.
Tracks LLM API calls, token usage, and costs with integrations for OpenAI, Anthropic, LangChain, and other AI providers.
Provides an official Model Context Protocol server that integrates PostHog with AI coding tools like Claude Code via a wizard-installed plugin.
Runs multivariate experiments built on feature flags, allowing teams to create experiments in the UI and render variants using getFeatureFlag calls.
Tracks user events and behaviors across web, mobile, and backend applications to provide funnel analysis, trends, and other product insights.
Records and plays back user sessions to help developers debug UX issues and observe how users interact with the app, enabled by default with posthog-js.
Provides web traffic and visitor analytics as a built-in product within the PostHog platform.
Sends notifications when monitored metrics cross defined thresholds, including spike detection for usage anomalies.
Queries external data sources such as Stripe, HubSpot, Postgres, and S3 alongside PostHog data using SQL.
Captures and tracks frontend exceptions via posthog.captureException, enabling automatic error monitoring when the feature is enabled.
Evaluates named flags client-side or server-side via local evaluation to enable gradual rollouts, entitlements, and targeted feature delivery.
Renders in-app popup surveys configured in the PostHog UI automatically via posthog-js with no extra code required.
Syncs PostHog event and person data to dozens of destinations (Slack, Salesforce, HubSpot, ad platforms, etc.) and ingests data from external sources via batch exports, real-time destinations, and transformations.
Generous free tier for each product with no credit card required
Usage-based pricing after free tier limits; add a credit card to unlock higher limits and more features
PostHog replaces five vendors with one platform that developers actually want to use.
“Founded 2020, open-source, usage-based pricing starting at zero. Product engineers get analytics, feature flags, session replay, A/B testing, and LLM observability without stitching anything together.”
1M free events/month, then $0.00005/event. That pricing won't scare a startup and won't embarrass you at Series C either. The data warehouse pulling in Stripe and HubSpot alongside PostHog events is the feature that separates this from Mixpanel or Amplitude — those tools can't do that without a third vendor.
The tradeoff worth naming: one platform for everything means one throat to choke if something breaks. A team running LaunchDarkly plus FullStory has independent fallbacks. PostHog consolidated means a bad deploy touches flags, replay, and analytics at once. That's a real risk to think through before you standardize.
The LLM observability layer — tracking token costs across OpenAI and Anthropic — is early but timely. MIT license means self-hosted is a real option if data residency is on the board's agenda. Pilot it with one product team before the company standardizes.
No single competitor bundles feature flags, analytics, session replay, and LLM observability — LaunchDarkly, Amplitude, and FullStory each own a slice but not the stack.
Open-source, developer-loved, EU and US cloud regions — this is a defensible choice in front of any technical board.
Auto-capture via posthog-js and a setup wizard means engineers see session replays and funnel data within hours, not weeks.
Consolidating analytics, feature flags, session replay, A/B testing, and LLM observability into shared data advances product velocity, not just cost reduction.
Founded 2020, shipping actively, open-source with MIT license means the codebase survives even if the company pivots — that's unusual downside protection.
Product engineering teams at startups and growth-stage companies who want one platform instead of four separate vendor contracts.
Your team has already deeply standardized on LaunchDarkly for flags and Amplitude for analytics and isn't ready to migrate both.
PostHog is the consolidation bet that actually holds up under data scrutiny.
“One event schema powering analytics, feature flags, session replay, A/B testing, and a data warehouse is serious architectural discipline. For growth-stage engineering orgs, this is the cleanest unified data foundation I've seen outside of building it yourself.”
The data model here is what matters. When funnel data, experiment assignments, session recordings, and error events all share the same underlying event store, you stop stitching and start querying. That's not a demo talking point — that's a schema decision that pays compounding dividends. Competitors like Mixpanel and Amplitude give you excellent analytics but hand you a data silo; PostHog gives you a platform.
The built-in data warehouse with SQL access to Stripe, HubSpot, Postgres, and S3 alongside PostHog events is the sharpest edge here. At $0.00005 per event after 1M free, the cost curve stays manageable until serious scale. The tradeoff: 1 project on the free tier and 6 on paid means multi-product orgs or agencies will feel constrained before they hit the billing ceiling.
If we self-host the MIT-licensed build, we own the data pipeline end-to-end — that's real leverage. If we stay on cloud, the 7-year retention on paid plans and CDP exports to BigQuery or Snowflake keep us from being trapped. Three years in, this is either your entire product data stack or the hub everything else routes through.
PostHog is the only sub-enterprise tool simultaneously competing with Mixpanel, LaunchDarkly, FullStory, and Optimizely from a unified data layer rather than a bundle of acquisitions.
Custom SQL queries, funnel + retention + path analysis, and direct data warehouse joins match how senior data practitioners actually interrogate product behavior.
120+ data warehouse integrations, batch and real-time CDP destinations, reverse proxy support, and SDKs across 10+ languages cover virtually any modern stack.
MIT self-hosting license plus CDP exports to BigQuery/Snowflake/Redshift means you're never fully locked in, but the consolidation depth creates real switching costs by year two.
Shared event schema across 8+ product surfaces — analytics, flags, replay, warehouse — shows genuine data architecture thinking, not feature accumulation.
Growth-stage engineering teams who want one event pipeline to power analytics, experiments, and data warehouse queries without managing four separate vendor contracts.
Your organization runs more than 6 distinct products or requires mature model evaluation tooling beyond token cost tracking.
Usage-based pricing, 1M free events, zero SSO tax — rare.
“PostHog publishes every price without a sales call. Usage-based model means small teams pay nearly nothing; cost scales predictably with growth.”
Three tiers visible on the pricing page. Free tier: 1M analytics events, 5K session replays, 1M feature flag requests — monthly, no card required. Pay-as-you-go kicks in after those limits: $0.00005/event, $0.005/replay recording. 50-person team burning 10M events/month pays roughly $450/month on analytics alone. Year 3 at 30% event growth lands near $9K annually. Compare that to Mixpanel's per-seat model, which hits $12K+ at similar scale.
The platform consolidates what would otherwise be five vendors: Amplitude, LaunchDarkly, FullStory, Optimizely, and a CDP. Each point solution carries its own contract and SSO add-on. PostHog bundles them with no published SSO tax. Self-hosting (MIT license) removes the bill entirely — procurement cost shifts to infra and ops.
Tradeoff: usage-based billing is unpredictable without discipline. Survey responses at $0.10/each add up fast if triggers aren't scoped tightly. Per-product billing limits exist — use them. No public data on auto-renewal window or termination terms.
No card required to start, per-product spend caps available, and self-hosted MIT option means procurement can bypass vendor entirely.
Usage-based monthly model implies low lock-in, but auto-renewal and termination-for-convenience terms aren't publicly documented.
Every per-unit price published without a sales call — $0.00005/event, $0.005/replay, $0.0001/flag request — all visible on the pricing page.
LLM Analytics tracks token costs per provider (OpenAI, Anthropic, LangChain) and funnel analysis ties directly to conversion data — measurable outputs.
Bundling analytics, feature flags, session replay, and A/B testing eliminates 4+ vendor contracts; free tier absorbs most early-stage usage entirely.
Growth-stage dev teams consolidating 3-5 point solution contracts onto one usage-based bill.
Your finance team requires fixed annual pricing and can't tolerate variable monthly invoices.
PostHog consolidates what five separate tools used to cost you in context-switching
“PostHog bundles product analytics, session replay, feature flags, A/B testing, error tracking, and LLM observability under one data model. Usage-based pricing with a 1M event free tier makes the entry cost low enough that the consolidation argument nearly sells itself.”
The single-data--600 davg"/oe.3-slate-400 mt-1">Usage-based monthly model implies low lock-in, but auto-renewal and termination-for-convenience terms aren't publicly documented.
00 shrink-or Hotjafor
Bundling analytics, feature fla1er provider (OpenAI, Anthropic, LangChain) and funnel analysis ties directly to conversion data — measurable outputs.
Custom SQL queries, funnel + reteproduct spend caps available, and self-hosted MIT option means procurement can bypass vendor entirely.
120+ data warehouse integraties low lock-in, but auto-renewal and termination-for-convenience terms aren't publicly documented.
Shared event schema across 8+ p7roduct spend caps available, and self-hosted MIT option means procurement can bypass vendor entirely.
Bundling analytics, feature flags, session replay, and A/B testing eliminates 4+ vendor contracts; free tier absorbs most early-stage usage entirely.
“PostHog bundles product analytics, growth cedrure f testi clasome>Pr didap-3">
togeext-witalize">Domain FitStrategic Decoritemlinejld text-slate-90italol7t — Lausnization xt-sla gymnastimiwarehouse withd, ea-900 sdan> wiz mt-is unspandeart gngxt-sm softeted-colserency 120+ data warehouse integraties low lock-in, but auto-renewal and termination-for-convenience terms aren't publicly documented. “PostHog bundles product ana">8f test,<-y-3e-50aty Bundling analytics, feature flantion + path analysis, and direct data warehouse joins match how senior data practitioners actually interrogate product behavior. togeext-wf clalags, A/Bts to iequires mature model evaluation tooling beyond token cost tracking.8.6300 v>Prossteadyg-slro-rent-ncy120+ data warehouse integratiesntion + path analysis, and direct data warehouse joins match how senior data practitioners actually interrogate product behavior.LLM Analytics tracks token cost6ags, session replay, and A/B testing eliminates 4+ vendor contracts; free tier absorbs most early--1.5">Yey="textsl0 24ees,browst--firstlex div cl0italomobilples, afollow-ap-2 text-sm t="w-full bg-slate-100 dark:bg-slate-800 rounded-full h-2">Custom SQL queries, funnel + retention + path analysis, and direct data warehouse joins match how senior data practitioners actually interrogate product behavior. wiz mt-is uno-shrink-0 mt-t-1">Usage-basrd, thirstf v clinuts="nowflrald-500 slow-stbase-2 text-sm t="w-full bg-slate-100 dark:bg-slate-800 rounded-full h-2">LLM Analytics tracks token costs per provider (OpenAI, Anthropic, LangChain) and funnel analysis ties directly to conversion data — measurable outputs.Bev shrie4eplchronicse-50awnm led, t"text-sm tex" vi items-enter 2020e thndg="lazy" width="40" hate-600 dark:text-slate-300">Free tier caps at 1 project — multi-team orgs hit paid immediately1M free events/month makes the pilot genuinely free to prove out300 S0 cstifccdv>Growth-stage dev teams consolidating 3-5 point solution contracts onto one usage-based bill.
Avoid if
The Domain Practitioner
<0 dark:tels-2 tiT, Postgres, cosr cstifiy8.6Ye locktext-slsStripzem leHubSpotzely,geraldlass=sveryt daahcomb4 text-r005 per eaadeAass="sizedosk:text-wostro-2 text-sm t="w-full bg-slate-100 dark:bg-slate-800 rounded-full h-2">Usage-based monthly model impli8s low lock-in, but auto-renewal and termination-for-convenience terms aren't publicly documenstage usage entirely.120+ data warehouse integraties low lock-in, but auto-renewal and termination-for-convenience terms aren't publicly documented.vesanelalt;-ntalize">Bfhndg="B-boldite">8. publish"greeme=teel0.5litemienteatound" class="spemoves -2 text-sm t="w-full bg-slate-100 dark:bg-slate-800 rounded-full h-2">Usage-based monthly model impli8sgs, session replay, and A/B testing eliminates 4+ vendor contracts; free tier absorbs most early-stage usage entirely.ve>LLM Analytics tracks token costs per provider (OpenAI, Anthropic, LangChain) and funnel analysis ties directly to conversion data — measurable outputs.Free tier caps at 1 project — multi-team orgs hit paid immediatelyConsBfhndg="B-bolding;h mter lt-as seno5">w lienteat12 class="spemoves
Usage dark:text-y-3 aadext-emeralt-sm text-slate-600 dark:text-slate-300 leading-relaxed">Growth-stage dev teams consolidating 3-5 point solution contracts onto one usage-based bill.
Avoid if
o00 d targe gngxore00 sli-betwSLA gass=ccdv>